David Vos's Avatar

David Vos

@davidvos.bsky.social

Responsible #RecSys and #IR in the Generative AI era. PhD Candidate at IRLab Amsterdam, supervised by Maarten de Rijke and Andrew Yates. davidvos.dev

601 Followers  |  304 Following  |  27 Posts  |  Joined: 08.11.2024  |  2.1939

Latest posts by davidvos.bsky.social on Bluesky

Let’s collaborate on democratizing insights from tabular data in Amsterdam! ✨

PhD directions: 1) fundamental techniques for tabular foundation models, 2) reliable mechanisms for AI-powered tabular data analysis.

Sharing w/ friends appreciated! ⬇️

05.06.2025 15:36 — 👍 2    🔁 3    💬 1    📌 0

The winner is '.translatesAutoresizingMaskIntoConstraints', with 42 characters in a single token. OpenAI really values iOS development I guess 😅

27.02.2025 11:46 — 👍 1    🔁 0    💬 0    📌 0
Post image

As it turns out: It's one of the longest tokens without special characters. When considering special characters, the top ones look like this.

27.02.2025 11:46 — 👍 0    🔁 0    💬 1    📌 0
Post image

Today's funny tokenization realization: The GPT4 tokenizer has 1 single token for the entire lower- and uppercase alphabet.

27.02.2025 11:29 — 👍 0    🔁 0    💬 1    📌 0

IRLab Amsterdam made it to Bluesky! Go give @irlab-amsterdam.bsky.social a follow for all things RecSys, IR, RAG, Conv AI, etc!

10.02.2025 15:06 — 👍 7    🔁 2    💬 0    📌 0
Post image

I'll get straight to the point.

We trained 2 new models. Like BERT, but modern. ModernBERT.

Not some hypey GenAI thing, but a proper workhorse model, for retrieval, classification, etc. Real practical stuff.

It's much faster, more accurate, longer context, and more useful. 🧵

19.12.2024 16:45 — 👍 620    🔁 147    💬 19    📌 34
Post image

Congratulations dr. Zihan Wang! It was an honor to be your paranymph.

13.12.2024 16:16 — 👍 9    🔁 0    💬 0    📌 0
PhD Position in Mechanistic Interpretability PhD Position in Mechanistic Interpretability

🚨 PhD position alert! 🚨

I'm hiring a fully funded PhD student to work on mechanistic interpretability at @uva-amsterdam.bsky.social. If you're interested in reverse engineering modern deep learning architectures, please apply: vacatures.uva.nl/UvA/job/PhD-...

02.12.2024 19:36 — 👍 21    🔁 11    💬 0    📌 1

I just completed "Historian Hysteria" - Day 1 - Advent of Code 2024 #AdventOfCode adventofcode.com/2024/day/1

01.12.2024 09:45 — 👍 2    🔁 0    💬 0    📌 0
Post image Post image

Yesterday @ellisamsterdam.bsky.social hosted the yearly NeurIPS-Fest, a pre-party for NeurIPS with a keynote talk, poster session, drinks and bites! 🍺🍻

The keynote was by @canaesseth.bsky.social , who talked about "Diffusion, Flows and other stories", presenting his 5 papers accepted at NeurIPS! 💥

29.11.2024 13:38 — 👍 27    🔁 7    💬 1    📌 0
Post image

I'm looking for an intern to introduce Sparse Embedding models to Sentence Transformers! If you're passionate about open source, interested in helping practitioners use your tools, and enjoy embedders/retrievers/rerankers, then I'd love to hear from you!

Links with details and to apply in 🧵

27.11.2024 14:31 — 👍 30    🔁 5    💬 3    📌 0
Evaluation Perspectives of Recommender Systems: Driving Research and Education (Dagstuhl Seminar 24211)

And we have a #Dagstuhl Report!
Evaluation Perspectives of #RecSys. Edited by @christinebauer.bsky.social, @evazangerle.bsky.social, and myself. Written by a whole host of fantastic #recsys people. Too many to mention (pic here www.dagstuhl.de/en/seminars/...)
drops.dagstuhl.de/entities/doc...

26.11.2024 20:57 — 👍 10    🔁 7    💬 0    📌 2
“Probabilistic machine learning”: a book series by Kevin Murphy “Probabilistic Machine Learning” - a book series by Kevin Murphy

I think it would be hard to beat the Probabilistic Machine Learning books by Kevin Murphy. probml.github.io/pml-book/

25.11.2024 08:50 — 👍 6    🔁 1    💬 2    📌 0

I love the formatting of this one :) Thank you!

25.11.2024 10:39 — 👍 1    🔁 0    💬 0    📌 0

Awesome, thank you!

25.11.2024 10:37 — 👍 1    🔁 0    💬 0    📌 0

I'm looking for a textbook to take me through fundamental ML concepts again. Not too applied, and preferably with a Deep Learning angle. Currently considering the new Deep Learning book by Bishop, but maybe my Bluesky following has different recommendations. Let me know :)

25.11.2024 08:31 — 👍 4    🔁 0    💬 2    📌 0
Every Academic Needs a Website

Things you need as a PhD student:
- coffee ☕
- a living stipend 💰
- a website 🔗

For the last one, @kiragoldner.bsky.social has good resources: www.kiragoldner.com/blog/website..., www.kiragoldner.com/resources.html

23.11.2024 05:29 — 👍 92    🔁 17    💬 1    📌 1
Preview
a panda bear is rolling around in the grass in a zoo enclosure . Alt: a panda bear is rolling around in the grass in a zoo enclosure .

No one can explain stochastic gradient descent better than this panda.

24.11.2024 15:04 — 👍 216    🔁 32    💬 10    📌 6

👋

24.11.2024 20:37 — 👍 0    🔁 0    💬 0    📌 0

👋

24.11.2024 20:30 — 👍 1    🔁 0    💬 0    📌 0
Preview
Top Information Retrieval Papers of the Week | Sumit | Substack A weekly curated newsletter about the latest research papers in the Information Retrieval domain, including Recommender Systems, Search, Retrieval, and Ranking. Click to read Top Information Retrieval...

This is such a great start of every week for me.

Credits to Sumit for compiling such an easy to digest newsletter on #IR, #RAG and #RecSys.

open.substack.com/pub/recsys?r...

24.11.2024 19:42 — 👍 3    🔁 2    💬 0    📌 0

Creating a 🦋 starter pack for people working in IR/RAG: go.bsky.app/88ULgwY

I can’t seem to find everyone though, help definitely appreciated to fill this out (DM or comment)!

23.11.2024 21:19 — 👍 86    🔁 23    💬 32    📌 1

Would love to be added! Thanks!

24.11.2024 08:16 — 👍 1    🔁 0    💬 0    📌 0

Would love to be added ✋

22.11.2024 15:12 — 👍 1    🔁 0    💬 0    📌 0
Post image

De overstap #BlueSky

22.11.2024 10:26 — 👍 1260    🔁 232    💬 36    📌 19

As people are slowly moving over, I thought I'd make a list of people on here who work on IR & RecSys in Amsterdam. Who am I missing? Both researchers and engineers :)

go.bsky.app/BzPgLK2

19.11.2024 13:42 — 👍 6    🔁 2    💬 3    📌 0

Haha all good! You should definitely come pay us a visit at some point :)

19.11.2024 15:47 — 👍 1    🔁 0    💬 0    📌 0

✅✅

19.11.2024 15:09 — 👍 0    🔁 0    💬 0    📌 0

Added you! Will keep you to that promise :)

19.11.2024 13:53 — 👍 1    🔁 0    💬 0    📌 0

As people are slowly moving over, I thought I'd make a list of people on here who work on IR & RecSys in Amsterdam. Who am I missing? Both researchers and engineers :)

go.bsky.app/BzPgLK2

19.11.2024 13:42 — 👍 6    🔁 2    💬 3    📌 0

@davidvos is following 20 prominent accounts